BotScreen is a fully distributed AI system that can detect aimbots in FPS games in a scalable and reliable manner using a trusted execution environment. The research team won the Distinguished Paper Award at the USENIX Security Symposium 2023   Cheating in online games poses significant threats to the game industry.  Cheaters can directly impact the revenue of game publishers by annoying benign pl...

  The softening IV needle can soften by body temperature upon insertion, deform and freely follow the motion of the peripheral vein without the risk of mechanical damage. Throughout its use, the needle remains entirely soft, preventing accidental needlestick injury and reuse. Medical IV needles are manufactured using hard materials such as stainless steel or plastic, which do not mechanically ma...

Voltage imaging, a technique for the optical measurement of electrical signals in neurons, shows promise for the future of neuroscience but is currently hindered by a low signal-to-noise ratio. SUPPORT, a newly developed self-supervised algorithm, overcomes this challenge by learning to denoise directly from noisy datasets.   Voltage imaging is a cutting-edge technique used in neuroscience to optically mea...

 Ghost imaging-based X-ray Fluorescence enables fast elemental mapping on the surface of a sample. However, achieving accurate element distributions still necessitates a significant amount of measured data. This research demonstrates Bayesian approaches for reconstructing high-quality elemental distributions with less data.  X-ray Fluorescence (XRF) analysis, a non-destructive inspection method, is a w...

 An improved discriminator guidance is provided to enhance the image generation quality of pre-trained diffusion models. This guidance constitutes a new diffusion model in combination of prior diffusion training and discriminator response.     The diffusion model has recently been highlighted for its success in image generation, video generation, and text-to-image generation. The state-of-the-art ...

Much of the research effort in robot manipulation is on enabling pick-and-place. However, not all objects can be grasped - some objects may be too big, thin, or heavy . In such cases, the robot needs to utilize non-grasping manipulation motions, such as pushing, toppling, and dragging, to move the object to the target location. Deep reinforcement learning algorithms are used to enable the robot with such capabili...